How are the alerts "bad link detected" and "communication link down" generated?

book

Article ID: 35469

calendar_today

Updated On:

Products

CA Spectrum

Issue/Introduction

This TecDoc describes how alerts such as "Bad Link Detected" and "Communication Link Down" are generated

How are the alerts “bad link detected” and “communication link down” generated?

 

Environment

Release: Spectrum 10.4.x
Component: SPCAEM - Events and Alarms

Resolution

Bad link Detected:
Every time the ‘Internal_link_status’ attribute is read (by a user from the UI, or CLI, or by code, stemming from polling, or fault isolation, traps, or even from neighbor devices interested in the attribute, etc) then we recalculate its value.

If we find that the new link status is not ‘Good’, we will check the ‘GeneratePortStatusAlarms’ (0x12a54) attribute, and, if set to ‘Yes’, we will generate the link event (and alarm) on the port model. So the bad link alarm may stem from any of the reasons above, simply reading the internal_link_status from the UI can cause it.

Setting the ‘GeneratePortStatusAlarms’ attribute to ‘No’ is the recommended setting if you don’t ever want to see such alarms on that port.
Bad link detected is generated via Spectrum´s fault intelligence.


Communication link is down:

This means an there is an unstable link such as a flapping port on a device.

The attribute ‘AssertLinkDownAlarm‘ (0x12957) attribute on the interface determines if this is generated and this attribute can be verified by both polling and trap.